CONTROL OF RNA FUNCTION BY CONFORMATIONAL DESIGN CYCLING IN ENERGY - PowerPoint PPT Presentation

CONTROL OF RNA FUNCTION BY CONFORMATIONAL DESIGN CYCLING IN ENERGY AND DESIGN LANDSCAPES Stefan Badelt Department of Theoretical Chemistry Theoretical Biochemistry Group (tbi) University of Vienna March 1 , 2016 st 1

OUTLINE RNA modeling and RNA design Design of RNAs with prion-like behavior Design of self-processing ribozymes Kinetics of RNA-RNA interactions 2

RNA STRUCTURE GCGGAUUUAGCUCAGUUGGGAGAGCGCCAGACUGAAGAUCUGGAGGUCCUGUGUUCGAUCCACAGAAUUCGCACCA A C C 1 1 A G C 50 C G 70 G C 60 U G 60 70 U A U 20 C A A U C A G C A U A G U C G G U U U U G C U A A 10 50 10 U C C U C G G U G A G C G A G G G G C 20 G A C G 40 A U 30 G C 30 40 A U C A U G G A A A secondary structure is a list of base pairs, where: A base may participate in at most one base pair Base pairs must not cross (no pseudoknots) Only isosteric base-pairs (GC, AU, GU) are allowed. 3

THE NEAREST NEIGHBOR ENERGY MODEL A A U H A U H G A A I C G G U 20 I I C U C A U A C I I G A 10 U G C 30 H: Hairpin loop M U U M: Multi loop C G I: Interior loop I G C E: Exterior loop G G I G C C A E A U 5' 3' 5' 3' E ( s ) = e ( l ) ∑ l ∈ s 4

ENERGY LANDSCAPES An energy landscape is defined by Conformation space s ∈ Ω Neighborhood relation [Move set] M ( s ) Energy function E ( s ) o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 5

ENERGY LANDSCAPES UGCGACGUCCGACCUCGUUUACGCCAGUACCCCACUUCUCUUUG 0 kcal/mol free energy [kcal/mol] equilibrium Z partition function folding pathways suboptimal structure prediction MFE minimum free energy structure prediction − E ( s ) P ( s ) = e − E ( s )/ kT kT Z = ∑ s ∈ Ω e G = − kT ln Z Z 6

COMPUTATIONAL RNA DESIGN FROM STRUCTURE(S) TO SEQUENCE GCGGAUUUAGC 1 70 UCAGUUGGGAG 60 AGCGCCAGACU GAAGAUCUGGA 10 50 GGUCCUGUGUU 20 CGAUCCACAGA Φ(σ) 30 40 AUUCGCACCA The objective function can vary Φ ( σ ) structure is MFE of sequence, maximize probability of structure, ... 7

RNA DESIGN IS FORMALLY HARD, BUT EASY IN PRACTICE sequences minimum free energy structures mapping redundant, sensitive (bigger) (common motifs realized more often) ... simple adaptive walks usally lead to satisfactory solutions 8

DEPENDENCY GRAPHS FOR BISTABLE RNA DESIGN .(((.((.(((...))).)).))) (((...)))((...))((...)). C. Flamm, I.L. Hofacker, S. Maurer-Stroh, P.F. Stadler, and M. Zehl. Design of multi-stable RNA molecules. RNA, 7:254–265, 2001 9

ON THE FUNCTION OF RIBOSWITCHES Anti- Ligand Terminator ON promoter region protein coding region riboswitch + Ligand - Ligand Terminator OFF promoter region protein coding region riboswitch 10

CAN WE DEMONSTRATE AUTOCATALTYIC COFORMATIONAL SELF-REPLICATION? 11

THE MECHANISM OF A PRION PrP + N N C C PrP C PrP C + N N C C N N N N PrP Sc -PrP Sc dimer PrP C PrP Sc PrP C -PrP Sc PrP Sc oligomer heterodimer PrP Sc protofibril PrP Sc oligomer N PrP Sc N PrP Sc N PrP Sc Aguzzi, A., Sigurdson, C., and Heikenwaelder, M. (2008). Molecular mechanisms of prion pathogenesis. Annual Review of Pathology: Mechanisms of Disease, 3, 11–40. 12

Requirements for an RNA prion Energy Landscape maximize refolding barrier free energy [kcal/mol] S 1 S 2 ΔG ‡ S 2 S 1 normal S 2 S 1 infectious MFE S 1 S 2 S 2 S 2 minimize refolding barrier + + free energy [kcal/mol] ΔG ‡ HIV Dis type S 1 S 2 S 2 S 2 kissing loop complex S 1 S 2 MFE S 2 S 2 S 2 S 2 S 1 S 2 13

PRION DESIGN PIPELINE ....(((((((..((((...(((((...)))))...))))..))))))) (((((((.........)))))))....((((((.........)))))). NNNNNNNAACCGACGANNNNNNNNNNNNNNNNNAACGUCGGANNNNNNN [switch.pl] thermodynamic candidate molecules [RNAsubopt + barriers] (M) [findpath] (H) [RNAsubopt + barriers] (D) final ranking of molecules 14

free energy [kcal/mol] ΔG ‡ S 1 +S 2 S 1 S 2 E(S 1 ) + E(S 2 ) ΔG ‡ S 1 S 2 S 2 S 2 E(S 1 S 2 ) E(S 2 S 2 ) 15

Z M Z c1 Z c2 Z dup Z S1 Z S2 Z S 1 Z S 1+ c 1 Z S 1+ c 2 [ S 1] = [ M ] + ( + ) [ D ] Z M Z c 1 Z c 2 16

FOLDING BARRIERS WITH AND WITHOUT AUTO-CATALYSIS S. 6.00 kcal/mol 5 S2 S2 => S1: S1 0 16.70 kcal/mol -5 -10.70 kcal/mol -12.70 kcal/mol free energy [kcal/mol] -10 0 20 40 60 80 -17.00 kcal/mol -16.10 kcal/mol -22.00 kcal/mol S2 + S2 + kiss -20 Energy Model 1 S1 + S2 S1 => S2: Energy Model 2 -25 13.60 kcal/mol -33.60 kcal/mol -30 S1 => S2: -23.40 kcal/mol -35 -31.80 kcal/mol 9.80 kcal/mol -39.00 kcal/mol -29.70 kcal/mol 0 20 40 60 80 length of refolding path [base-pair moves] Badelt, C. Flamm, and I.L. Hofacker. Computational design of a circular RNA with prionlike behavior. Artificial Life 22, pages 1–14, 2016 17

CAN AN RNA CIRCULARIZE ITSELF? R 1 R 1 R 1 O B 1 O O O B 1 B 1 O O H :B - δ − O O B B H O O O O cleavage P H P O - O - O - O P O O - δ + ligation O O H H O B 2 A B 2 B 2 O A: O + A-H O O O O O O O H H H R 2 R 2 R 2 18

DESIGN OF SELF-PROCESSING RNA LCR L + CR L + C + R LCR LC + R L + O + R 19

Greifswald: Sabine Müller Biochemistry Sonja Petkovic Greifswald/Göteborg: RNA1 5' 92 3' 83 -29.80 -23.80 4.5 9.7 3.6 0.0 103 5' K 92 3' K 5'•92 83•3' Mihaela Delcea -31.10 (4.2) -31.40 -23.00 (7.2) -23.30 1/K 4.2 8.9 83 c c83 l -27.60 (3.1) -23.20 (2.8) 103 3' K 94 5' K 94•3' 5'•83 -26.40 (8.5) -26.70 -29.20 (2.8) -29.50 5.7 5.0 3.6 6.2 94 83 3' -27.20 5' -27.50 RNA2 5' 92 3' 83 -27.80 -28.00 7.3 3.5 7.7 0.9 103 5' K 5'•92 92 3' K 83•3' -34.00 -34.10 -33.90 (1.9) -34.00 (1.3) 1/K 9.7 9.6 83 c c83 l -31.60 (1,4) -27.40 (1.3) 103 3' K 94 5' K 94•3' 5'•83 -34.00 (1.6) -34.10 -33.90 (1.3) -34.00 7.3 3.5 7.7 3.7 3' 94 5' 83 Stephan Block -28.00 -27.80 Physics Wien: Ivo Hofacker Computational Christoph Flamm Biology Stefan Badelt 20

LCR L + CR L + C + R LCR LC + R L + O + R compute a set of candidate molecules (switch.pl) maximize probabilities to form reactive conformations optimize towards reactive monomers and/or dimers ***** * * * ****** ******* ********* ***** CRZ2 GGGAGAUCACAGUCCUCUUUGACGGGGUUCCGUCAAAGAGAGAAGUGAACCAGAGAAACACACUUCGGUGGUAUAUUACCUGGUCCCCCUCACAGUCCUCUUU---- PBD1 GGGAGAGCACAGUCGGAGUUGCCGCGUUAGCGGCGGUUCUAGAAGUGCCCCGCAGAAACAGCCAUAUGGCGUAUAUUACGCGGGAAAAAGCACAGUCGGAACC---- PBD2 GGGAGAGAACAGUCGGUGGUGCCCCGUAAGGGGCGUCGCCAGAAGUUCGGACCAGAAACAGCCAAAAGGCGUAUAUUACGGUCCAAAAAGAACAGUCGGCGAC---- --- GGGAGA CAGUCCGGUUUACCGCUAAUGCGGUGGGUCGAGAAGUCUGAGCGAGAAACACAGUAUACUGGUAUAUUACCGCUCCAUAAAGGCAGUCCGGCACCAAA PBD3 PBD4 --- GGGAGA CAGUCCGGUUUACCGCUAAUGCGGUGGGUCGAGAAGUCUGAGCGAGAAACACAGGACACUGGUAUAUUACCGCUCCAUAAAGGCAGUCCGGCACCAAA ---...((((....(((((((((((....)))))))))))....))))(((((.......(((....))).........)))))........................ ---.................(((((....)))))((((((....(((((((((.......(((....))).........))))).....))))....))))))).... 1.......10........20........30........40........50........60........70........80........90.......100........ 21

2D 3D AFM 22

DISSOCIATION BARRIERS DETERMINE EFFICIENCY free energy (activation energy) 5' 92 3' 83 CRZ-2 -29.80 -23.80 4.5 9.7 3.6 0.0 (2.7) 103 5' 92 3' 5'•92 83•3' -31.10 (4.2) -31.40 -23.00 (7.2) -23.30 83 c c83 l 4.2 8.9 -27.60 (3.1) -23.20 (2.8) 103 3' 94 5' 94•3' 5'•83 -26.40 (8.5) -26.70 -29.20 (2.8) -29.50 5.7 5.0 3.6 6.2 94 83 3' 5' -27.20 -27.50 refolding barrier dissociation barrier 23

RESULTS RNA (in theory) capable of conformational self-replication ribozymes can be designed to ligate themselves energy barriers for dissociation are important for design experimental data to model interactions of ribozymes 24

25 design final observation Input Output modeling SUMMARY AGCAACA AGGAUCU AGGAUCA AGGAACA final state 2.6 0.8 3.1 1 0.7 1.2 0.8 12 10 1.2 1.7 34 29 0.8 1.1 41 0.7 0.599999 1.0 80 73 91 2.6 43 Design landscapes Energy landscapes 2.3 2.3 100 90 3.9 4.4 42 68 0.8 3.1 2 0.8 13 11 0.6 2.6 2.6 6 0.9 21 7 0.6 0.8 20 19 0.8 0.8 26 0.8 0.8 25 24 0.8 47 48 0.8 0.8 1.5 67 66 3.0 46 1.0 62 18 2.0 1.3 1.0 27 22 modeling 1.4 71 1.3 Reality 2.0 1.0 28 23 design 0.7 99 1.3 69 2.8 1.0 75 2.4 2.6 61 50 0.8 2.3 44 36 2.0 2.3 2.9 64 1.1 2.6 56 35 3.0 76 49 2.8 2.8 98 82 2.2 0.7 1.4 3 1.2 3.5 8 4 0.799999 40 1.2 16 0.8 1.2 33 1.2 45 37 A G A U 0.7 0.8 57 53 A U A 0.7 65 3.1 U A G G 1.3 0.8 70 17 G U 84 58 C G C 3.5 15 C G G 87 1.9 30 C G 92 32 C A U A 1.6 60 4.2 14 C C G 0.7 1.5 1.9 63 U G C A U G U C 1.3 78 54 A U C U A A 1.2 1.3 96 89 U U C U A 2.3 1.2 97 79 1.4 3.5 5 0.8 51 2.4 9 0.8 1.2 52 1.3 85 39 1.3 0.8 81 59 0.8 1.2 83 1.9 74 0.7 31 93 38 2.6 1.5 88 72 2.6 4.2 86 3.5 77 55 3.9 3.7 95 94 -8.0 -10.0 -12.0 -14.0 -16.0 -18.0 initial state free energy objective function AGGAACAGUCGCACU ACCCACCUCGACAUC GUAAAUCAAAUUGGA ACUGAAGCCCUUGGU CUGGAGUCACCAGGG GGUUUACGUACUACU AGGAACAGUCGCACU ACCCACCUCGACAUC GUAAAUCAAAUUGGA ACUGAAGCCCUUGGU CUGGAGUCACCAGGG GGUUUACGUACUACU design experimental setup modeling >Output >Input